Impact Evaluation of a School-Based Sexuality and HIV Prevention Education Activity in South Africa

The purpose of this study is to evaluate the impact of the new South African Department of Basic Education (DBE) sexuality and HIV education program that uses scripted lesson plans and supporting activities and will be implemented by Education Development Center (EDC). The primary outcomes are the incidence of HSV-2 or pregnancy among a cohort of learners in grade 8 and followed for two years in two provinces of South Africa.

This evaluation focuses on the HIV and sexual and reproductive health (SRH) content of the Government of South Africa life orientation (LO) Curriculum. In 2010, the Department of Basic Education (DBE) undertook assessments of their LO program and learned that while young people who participated in the program had improved knowledge and attitudes, the program was not being implemented uniformly, such that the results were inconsistent across schools and learners. To address these concerns, DBE, with support from the United States Agency for International Development (USAID) and the United States President's Emergency Plan for AIDS Relief (PEPFAR), developed scripted lesson plans (SLP) to strengthen the SRH content and standardize implementation across schools; these SLP were paired with supporting activities to address fidelity in the curriculum's delivery. The SLP were integrated in the LO program, with most of the lessons offered in the first half of the school year. The lessons were developed for grades 4-6, 7-9, and 10-12. This evaluation focuses on the curricula for grades 7-9 and 10-12. There are eight lessons for grade 7, eight for grade 8, 11 for grade 9, and 10 for grade 10. Notably, there were delays in getting DBE approvals for release of the grade-10 curriculum. A key component of the program is that all LO teachers are trained on the new materials prior to implementation. Since 2015, USAID/PEPFAR has provided technical support to DBE for program roll out in priority provinces and districts that have the highest HIV incidence and prevalence. Technical support includes educator training on the new SLP and considerations of approaches for scale-up beyond the initial districts. Support for the first phase of implementation and testing was led by Education Development Center, with funding from USAID. The goal of the proposed impact evaluation is to assess the impact of the LO program on students over time. The primary evaluation question is: What is the effect of the scripted lesson plans and supporting activities on the incidence of HSV-2 or pregnancy after two years among a cohort of girls enrolled in grade 8 at intervention schools compared to a cohort of girls in grade 8 at control schools providing the current life skills program (i.e., the standard of practice)? The secondary evaluation questions include: What is the effect of the scripted lesson plans and supporting activities on knowledge, attitudes, school retention, and self-reported risk behavior, HIV testing, and completed referrals for health services at the end of 8th, 9th, and 10th grade among a cohort of girls first interviewed in grade 8 and among a cross-section girls and boys interviewed in grade 8, grade 9 and grade 10? If there is a reduction in the primary outcome, does the intervention work by increasing school retention, or is the effect independent of school retention?

Randomized cluster design whereby schools were assigned to intervention or control schools before the trial began.

Participants are either in intervention or control schools (intervention schools received the new curriculum); participants did not know about this prior assignment.

Participants in this arm were in schools where the life orientation teachers in grades 7-9 (in 2016 and 2017) and grade 10 in 2018 were trained to use the new life orientation curriculum that included scripted lesson plans for the sexual and reproductive health content of the program. There are eight lessons for grade 7, eight for grade 8, 11 for grade 9, and 10 for grade 10.

Participants in this arm were in schools where the standard life orientation curriculum was used with no additional training and no use of the new materials.

The revised life orientation content and implementation approach were designed to address gaps in prior implementation of the sexuality and HIV education content. By using scripted lesson plans and training teachers to use these, it was thought that students would get better information and lessons on the material important for HIV prevention.

Numerator: number of girls in the longitudinal cohort that test positive for genital herpes (HSV-2); Denominator: all cohort girls who did not have HSV-2 at baseline HSV-2 is measured using biomarkers. Dried blood spots were collected at baseline and again and endline. The baseline dried blood spots were stored in freezers until after analysis of the endline dried blood spots. Those who tested positive for genital herpes at endline were tested at baseline to determine if it was a new (incident) infection or if the girl already had genital herpes at baseline. The incidence measure is a change from her earlier status.

The percentage of girls in the longitudinal cohort who were never pregnant at baseline and who experience a pregnancy (self-reported) by endline. This is measuring a change in her experience with a pregnancy (incidence).

Percentage of girls in the longitudinal cohort who experience either incident HSV-2 infection or a pregnancy since baseline. This is measuring a change her her status.

Eight items were measured at each survey round to determine young people's knowledge of how HIV spreads. Each question was posed as a "True," "False," or "Don't know" option. Correct responses were coded one and incorrect or don't know responses were coded zero. Correct responses were summed so that a higher knowledge score indicates greater knowledge about HIV risk.

This was measured at each survey round using 17 items that were designed based on the Gender Equitable Men Scale, with additional items added. All questions were asked on a scale of "Agree a lot" to "Do not agree at all." These were coded as Agree a lot=1 to Do not agree at all =3. All were recoded so that higher values represent more equitable attitudes. We sum up the 17 items so that the summed scale equates to higher values being more equitable attitudes.

This was measured at each survey round by asking participants to self report if they visited a facility for an HIV test in the last 12 months. This is used to examine participants experience with HIV testing and if there are changes over time across the study groups.

Using dried blood spots collected in the female cross-sectional sample at endline as well as the female cohort, the investigators are able to measure the prevalence of HIV at endline. This was an outcome that was added in the course of the study and was not part of the original protocol.

Measured by self-reported number of sexual partners in the last year. This will be categorized once the distribution is examined.

Inclusion Criteria: For the main cohort sample - females in grade 8 were enrolled in 2016 and followed for two years. All participating girls had to receive parental consent, provide assent, and give contact information for study linking. For the cross-sectional samples, female and male learners in grade 8 in 2016; in grade 9 in 2017 and in grade 10 in 2018 were eligible to participate assuming they had parental consent and assented to participate. Exclusion Criteria: Must understand one of the five languages that the survey was administered in.

De-identified data are available on a publicly available server: https://dataverse.unc.edu/dataverse/cpc

Speizer IS, Mandal M, Xiong K, Hattori A, Makina-Zimalirana N, Kumalo F, Taylor S, Ndlovu MS, Madibane M, Beke A. Methodology and Baseline Results From the Evaluation of a Sexuality Education Activity in Mpumalanga and KwaZulu-Natal, South Africa. AIDS Educ Prev. 2018 Apr;30(2):152-168. doi: 10.1521/aeap.2018.30.2.152.

Speizer IS, Xiong K, Mandal M, Makina-Zimalirana N, Hattori A, Durno D. HIV-Related Knowledge, Attitudes, and Behaviors among Grade 10 Girls and Boys in Mpumalanga and KwaZulu-Natal: Cross-Sectional Results. Open AIDS J. 2020;14:75-83. doi: 10.2174/1874613602014010075. Epub 2020 Sep 18.

Speizer IS, Mandal M, Xiong K, Makina N, Hattori A, Durno D. Impact evaluation of scripted lesson plans for HIV-related content in a life orientation curriculum: results from two provinces in South Africa. BMC Public Health. 2020 Oct 14;20(1):1542. doi: 10.1186/s12889-020-09640-2.